Harsh Reality
•
Dynamic Memory Allocation
• Memory Matters
Memory is not unbounded
– It must be allocated and managed
– Many applications are memory dominated
• Especially those based on complex, graph algorithms
•
•
Memory referencing bugs especially pernicious
– Effects are distant in both time and space
Memory performance is not uniform
– Cache and virtual memory effects can greatly affect program
performance
– Adapting program to characteristics of memory system can lead
to major speed improvements
class12.ppt
Dynamic Memory Allocation
Process Memory Image
Application
kernel virtual memory
Dynamic Memory Allocator
memory invisible to
user code
stack
%esp
Heap Memory
•
Explicit vs. Implicit Memory Allocator
– Explicit: application allocates and frees space
• E.g., malloc and free in C
– Implicit: application allocates, but does not free space
• E.g. garbage collection in Java, ML or Lisp
•
Memory mapped region for
shared libraries
Allocators request
additional heap memory
from the operating
system using the sbrk
function.
the “brk” ptr
run-time heap (via malloc)
Allocation
– In both cases the memory allocator provides an abstraction of
memory as a set of blocks
– Doles out free memory blocks to application
uninitialized data (.bss)
initialized data (.data)
program text (.text)
0
Page 1
Malloc Package
Malloc Example
void foo(int n, int m) {
int i, *p;
•
•
#include <stdlib.h>
void *malloc(size_t size)
– If successful:
/* allocate a block of n ints */
if ((p = (int *) malloc(n * sizeof(int))) == NULL) {
perror("malloc");
exit(0);
}
for (i=0; i<n; i++)
p[i] = i;
• Returns a pointer to a memory block of at least size bytes, (typically)
aligned to 8-byte boundary.
• If size == 0, returns NULL
– If unsuccessful: returns NULL (0) and sets errno.
•
•
/* add m bytes to end of p block */
if ((p = (int *) realloc(p, (n+m) * sizeof(int))) == NULL)
{
void free(void *p)
– Returns the block pointed at by p to pool of available memory
– p must come from a previous call to malloc or realloc.
perror("realloc");
exit(0);
}
for (i=n; i < n+m; i++)
p[i] = i;
void *realloc(void *p, size_t size)
– Changes size of block p and returns pointer to new block.
– Contents of new block unchanged up to min of old and new size.
/* print new array */
for (i=0; i<n+m; i++)
printf("%d\n", p[i]);
free(p); /* return p to available memory pool */
}
Assumptions
Allocation Examples
p1 = malloc(4)
•
Assumptions made in this lecture
p2 = malloc(5)
– Memory is word addressed (each word can hold a pointer)
Allocated block
(4 words)
Free block
(3 words)
p3 = malloc(6)
Free word
Allocated word
free(p2)
p4 = malloc(2)
Page 2
Constraints
Goals of Good malloc/free
• Applications:
– Can issue arbitrary sequence of allocation and free requests
– Free requests must correspond to an allocated block
• Allocators
– Can’t control number or size of allocated blocks
– Must respond immediately to all allocation requests
•
Primary goals
– Good time performance for malloc and free
• Ideally should take constant time (not always possible)
• Should certainly not take linear time in the number of blocks
– Good space utilization
• User allocated structures should be large fraction of the heap.
• Want to minimize “fragmentation”.
• i.e., can’t reorder or buffer requests
– Must allocate blocks from free memory
•
• i.e., can only place allocated blocks in free memory
– Must align blocks so they satisfy all alignment requirements
Some other goals
– Good locality properties
• Structures allocated close in time should be close in space
• “Similar” objects should be allocated close in space
• 8 byte alignment for GNU malloc (libc malloc) on Linux boxes
– Can only manipulate and modify free memory
– Can’t move the allocated blocks once they are allocated
– Robust
• Can check that free(p1) is on a valid allocated object p1
• i.e., compaction is not allowed
• Can check that memory references are to allocated space
Performance Goals:
Peak Memory Utilization
Performance Goals: Throughput
•
Given some sequence of malloc and free requests:
– R0, R1, ..., Rk, ... , Rn-1
•
Given some sequence of malloc and free requests:
– R0, R1, ..., Rk, ... , Rn-1
•
Want to maximize throughput and peak memory utilization.
– These goals are often conflicting
•
•
Throughput:
– Number of completed requests per unit time
– Example:
•
Def: Aggregate payload Pk:
– malloc(p) results in a block with a payload of p bytes..
– After request Rk has completed, the aggregate payload Pk is the
sum of currently allocated payloads.
Def: Current heap size is denoted by Hk
– Assume that Hk is monotonically nondecreasing
Def: Peak memory utilization:
– After k requests, peak memory utilization is:
•
• 5,000 malloc calls and 5,000 free calls in 10 seconds
• Throughput is 1,000 operations/second.
• Uk = ( maxi<k Pi ) / Hk
Page 3
Internal Fragmentation
External Fragmentation
Occurs when there is enough aggregate heap memory, but no single
free block is large enough
•
•
Poor memory utilization caused by fragmentation.
– Comes in two forms: internal and external fragmentation
Internal fragmentation
– For some block, internal fragmentation is the difference between the
block size and the payload size.
p1 = malloc(4)
p2 = malloc(5)
block
Internal
fragmentation
p3 = malloc(6)
Internal
fragmentation
payload
free(p2)
– Caused by overhead of maintaining heap data structures, padding for
alignment purposes, or explicit policy decisions (e.g., not to split the
block).
– Depends only on the pattern of previous requests, and thus is easy to
measure.
p4 = malloc(6)
oops!
External fragmentation depends on the pattern of future requests, and
thus is difficult to measure.
Implementation Issues
Knowing How Much to Free
• Standard method
– Keep the length of a block in the word preceding the
block.
• How do we know how much memory to free just given a
pointer?
• How do we keep track of the free blocks?
• What do we do with the extra space when allocating a
structure that is smaller than the free block it is placed
in?
• How do we pick a block to use for allocation -- many
might fit?
• How do we reinsert freed block?
• This word is often called the header field or header
– Requires an extra word for every allocated block
p0 = malloc(4)
p0
5
p0
free(p0)
free(p0)
p1 = malloc(1)
Page 4
Block size
data
Keeping Track of Free Blocks
Method 1: Implicit List
• Need to identify whether each block is free or allocated
– Can use extra bit
– Bit can be put in the same word as the size if block
sizes are always multiples of two (mask out low order
bit when reading size).
• Method 1: Implicit list using lengths -- links all blocks
5
4
6
2
• Method 2: Explicit list among the free blocks using
pointers within the free blocks
1 word
5
4
6
2
size
• Method 3: Segregated free list
– Different free lists for different size classes
• Method 4: Blocks sorted by size
– Can use a balanced tree (e.g. Red-Black tree) with
pointers within each free block, and the length
used as a key
Format of
allocated and
free blocks
•
•
payload: application data
(allocated blocks only)
Implicit List: Allocating in Free Block
•
First fit:
– Search list from beginning, choose first free block that fits
p = start;
while ((p < end) ||
(*p & 1) ||
(*p <= len));
size: block size
payload
optional
padding
Implicit List: Finding a Free Block
•
a = 1: allocated block
a = 0: free block
a
\\ not passed end
\\ already allocated
\\ too small
Allocating in a free block - splitting
– Since allocated space might be smaller than free space, we
might want to split the block
4
4
6
2
p
– Can take linear time in total number of blocks (allocated and free)
– In practice it can cause “splinters” at beginning of list
Next fit:
– Like first-fit, but search list from location of end of previous search
– Research suggests that fragmentation is worse
Best fit:
– Search the list, choose the free block with the closest size that fits
– Keeps fragments small --- usually helps fragmentation
– Will typically run slower than first-fit
void addblock(ptr p, int len) {
int newsize = ((len + 1) >> 1) << 1;
int oldsize = *p & -2;
*p = newsize | 1;
if (newsize < oldsize)
*(p+newsize) = oldsize - newsize;
}
// add 1 and round up
// mask out low bit
// set new length
// set length in remaining
//
part of block
addblock(p, 2)
4
Page 5
4
4
2
2
Implicit List: Freeing a Block
•
Implicit List: Coalescing
• Join (coelesce) with next and/or previous block if they
are free
– Coalescing with next block
Simplest implementation:
– Only need to clear allocated flag
void free_block(ptr p) { *p = *p & -2}
– But can lead to “false fragmentation”
4
4
4
2
2
2
2
void free_block(ptr p) {
*p = *p & -2;
// clear allocated flag
next = p + *p;
// find next block
if ((*next & 1) == 0)
*p = *p + *next;
// add to this block if
}
//
not allocated
p
free(p)
4
malloc(5)
4
4
Oops!
4
4
free(p)
There is enough free space, but the allocator won’t be able to find it
4
2
2
p
4
4
6
2
– But how do we coalesce with previous block?
Implicit List: Bidirectional Coalescing
•
Constant Time Coalescing
Boundary tags [Knuth73]
– Replicate size/allocated word at bottom of free blocks
– Allows us to traverse the “list” backwards, but requires extra space
– Important and general technique!
1 word
Header
Format of
allocated and
free blocks
a
payload and
padding
Boundary tag
(footer)
4
size
4 4
size
4 6
block being
freed
a = 1: allocated block
a = 0: free block
size: total block size
a
payload: application data
(allocated blocks only)
6 4
4
Page 6
Case 1
Case 2
Case 3
Case 4
allocated
allocated
free
free
allocated
free
allocated
free
Constant Time Coalescing (Case 1)
Constant Time Coalescing (Case 2)
m1
1
m1
1
m1
1
m1
1
m1
n
1
1
m1
n
1
0
m1
n
1
1
m1
n+m2
1
0
n
m2
1
1
n
m2
0
1
n
m2
1
0
m2
1
m2
1
m2
0
n+m2
0
Constant Time Coalescing (Case 3)
m1
0
n+m1
m1
n
0
1
n
m2
1
1
n+m1
m2
m2
1
m2
Constant Time Coalescing (Case 4)
0
m1
0
m1
n
0
1
0
1
n
m2
1
0
1
m2
0
Page 7
n+m1+m2
0
n+m1+m2
0
Summary of Key Allocator Policies
•
Implicit Lists: Summary
Placement policy:
– First fit, next fit, best fit, etc.
– Trades off lower throughput for less fragmentation
• Interesting observation: segregated free lists (next lecture) approximate a
best fit placement policy without having the search entire free list.
•
•
Splitting policy:
– When do we go ahead and split free blocks?
– How much internal fragmentation are we willing to tolerate?
Coalescing policy:
– Immediate coalescing: coalesce adjacent blocks each time free is called
– Deferred coalescing: try to improve performance of free by deferring
coalescing until needed. e.g.,
•
•
•
•
Implementation: very simple
Allocate: linear time worst case
Free: constant time worst case -- even with coalescing
Memory usage: will depend on placement policy
– First fit, next fit or best fit
•
Not used in practice for malloc/free because of linear time allocate.
Used in many special purpose applications.
•
However, the concepts of splitting and boundary tag coalescing are
general to all allocators.
• Coalesce as you scan the free list for malloc.
• Coalesce when the amount of external fragmentation reaches some
threshold.
Keeping Track of Free Blocks
Explicit Free Lists
A
B
C
• Method 1: Implicit list using lengths -- links all blocks
5
4
6
2
•
• Method 2: Explicit list among the free blocks using
pointers within the free blocks
Use data space for link pointers
– Typically doubly linked
– Still need boundary tags for coalescing
Forward links
5
4
6
A
2
4
• Method 3: Segregated free lists
– Different free lists for different size classes
• Method 4: Blocks sorted by size (not discussed)
– Can use a balanced tree (e.g. Red-Black tree) with
pointers within each free block, and the length used
as a key
B
4 4
4 6
6 4
C
4 4
4
Back links
– It is important to realize that links are not necessarily in the same
order as the blocks
Page 8
Allocating From Explicit Free Lists
pred
Before:
Freeing With Explicit Free Lists
•
succ
free block
Insertion policy: Where in the free list do you put a newly freed
block?
– LIFO (last-in-first-out) policy
• Insert freed block at the beginning of the free list
• Pro: simple and constant time
• Con: studies suggest fragmentation is worse than address ordered.
pred
After:
(with splitting)
– Address-ordered policy
succ
• Insert freed blocks so that free list blocks are always in address
order
free block
– i.e. addr(pred) < addr(curr) < addr(succ)
• Con: requires search
• Pro: studies suggest fragmentation is better than LIFO
Freeing With a LIFO Policy (cont)
Freeing With a LIFO Policy
p
pred (p)
s
before:
succ (s)
f
•
Case 1: a-a-a
– Insert self at beginning of
free list
a
self
p
•
Case 2: a-a-f
– Splice out next, coalesce
self and next, and add to
beginning of free list
•
a
Case 3: f-a-a
– Splice out prev, coalesce
with self, and add to
beginning of free list
p
s
f
p
a
s1
p2
s2
before:
f
after:
a
s
p1
self
a
after:
before:
a
self
•
s
Case 4: f-a-f
– Splice out prev and next,
coalesce with self, and add
to beginning of list
f
p1
self
s1
p2
after:
f
f
Page 9
f
s2
Keeping Track of Free Blocks
Explicit List Summary
•
•
• Method 1: Implicit list using lengths -- links all blocks
Comparison to implicit list:
– Allocate is linear time in number of free blocks instead of total
blocks -- much faster allocates when most of the memory is full
– Slightly more complicated allocate and free since needs to splice
blocks in and out of the list
– Some extra space for the links (2 extra words needed for each
block)
Main use of linked lists is in conjunction with segregated free lists
– Keep multiple linked lists of different size classes, or possibly for
different types of objects
5
4
6
2
• Method 2: Explicit list among the free blocks using
pointers within the free blocks
5
4
6
2
• Method 3: Segregated free list
– Different free lists for different size classes
• Method 4: Blocks sorted by size
– Can use a balanced tree (e.g. Red-Black tree) with
pointers within each free block, and the length
used as a key
Segregated Storage
Simple Segregated Storage
•
•
•
• Each size class has its own collection of blocks
1-2
Separate heap and free list for each size class
No splitting
To allocate a block of size n:
– If free list for size n is not empty,
• allocate first block on list (note, list can be implicit or explicit)
3
– If free list is empty,
• get a new page
• create new free list from all blocks in page
• allocate first block on list
4
5-8
•
9-16
– Often have separate size class for every small size (2,3,4,…)
– For larger sizes typically have a size class for each power of 2
•
Page 10
– Constant time
To free a block:
– Add to free list
– If page is empty, return the page for use by another size (optional)
Tradeoffs:
– Fast, but can fragment badly
Segregated Fits
•
•
For More Info on Allocators
Array of free lists, each one for some size class
To allocate a block of size n:
– Search appropriate free list for block of size m > n
– If an appropriate block is found:
• D. Knuth, “The Art of Computer Programming, Second
Edition”, Addison Wesley, 1973
– The classic reference on dynamic storage allocation
• Split block and place fragment on appropriate list (optional)
•
•
– If no block is found, try next larger class
– Repeat until block is found
To free a block:
– Coalesce and place on appropriate list (optional)
Tradeoffs
– Faster search than sequential fits (i.e., log time for power of two
size classes)
– Controls fragmentation of simple segregated storage
– Coalescing can increase search times
• Wilson et al, “Dynamic Storage Allocation: A Survey and
Critical Review”, Proc. 1995 Int’l Workshop on Memory
Management, Kinross, Scotland, Sept, 1995.
– Comprehensive survey
– Available from CS:APP student site
(csapp.cs.cmu.edu)
• Deferred coalescing can help
Implicit Memory Management:
Garbage Collection
Garbage Collection
• How does the memory manager know when memory
can be freed?
– In general we cannot know what is going to be used in
the future since it depends on conditionals
– But we can tell that certain blocks cannot be used if
there are no pointers to them
• Garbage collection: automatic reclamation of heapallocated storage -- application never has to free
void foo() {
int *p = malloc(128);
return; /* p block is now garbage */
}
• Need to make certain assumptions about pointers
– Memory manager can distinguish pointers from nonpointers
– All pointers point to the start of a block
– Cannot hide pointers (e.g., by coercing them to an
int, and then back again)
• Common in functional languages, scripting languages,
and modern object oriented languages:
– Lisp, ML, Java, Perl, Mathematica,
• Variants (conservative garbage collectors) exist for C
and C++
– Cannot collect all garbage
Page 11
Memory as a Graph
Classical GC algorithms
•
• Mark and sweep collection (McCarthy, 1960)
– Does not move blocks (unless you also “compact”)
• Reference counting (Collins, 1960)
– Does not move blocks (not discussed)
• Copying collection (Minsky, 1963)
– Moves blocks (not discussed)
We view memory as a directed graph
– Each block is a node in the graph
– Each pointer is an edge in the graph
– Locations not in the heap that contain pointers into the heap are
called root nodes (e.g. registers, locations on the stack, global
variables)
Root nodes
reachable
Heap nodes
Not-reachable
(garbage)
• For more information, see Jones and Lin, “Garbage
Collection: Algorithms for Automatic Dynamic Memory”,
John Wiley & Sons, 1996.
•
•
A node (block) is reachable if there is a path from any root to that node.
Non-reachable nodes are garbage (never needed by the application)
Mark and Sweep Collecting
Assumptions For This Lecture
•
•
Application
– new(n): returns pointer to new block with all locations cleared
– read(b,i): read location i of block b into register
– write(b,i,v): write v into location i of block b
•
•
Each block will have a header word
– addressed as b[-1], for a block b
Can build on top of malloc/free package
– Allocate using malloc until you “run out of space”
When out of space:
– Use extra mark bit in the head of each block
– Mark: Start at roots and set mark bit on all reachable memory
– Sweep: Scan all blocks and free blocks that are not marked
– Used for different purposes in different collectors
•
Mark bit set
root
Before mark
Instructions used by the Garbage Collector
– is_ptr(p): determines whether p is a pointer
– length(b): returns the length of block b, not including the header
– get_roots(): returns all the roots
After mark
After sweep
Page 12
free
free
Conservative Mark and Sweep in
C
Mark and Sweep (cont.)
Mark using depth-first traversal of the memory graph
ptr mark(ptr p) {
if (!is_ptr(p)) return;
if (markBitSet(p)) return
setMarkBit(p);
for (i=0; i < length(p); i++)
mark(p[i]);
return;
}
•
//
//
//
//
do nothing if not pointer
check if already marked
set the mark bit
mark all children
A conservative collector for C programs
– Is_ptr() determines if a word is a pointer by checking if it
points to an allocated block of memory.
– But, in C pointers can point to the middle of a block.
ptr
header
Sweep using lengths to find next block
•
ptr sweep(ptr p, ptr end) {
while (p < end) {
if markBitSet(p)
clearMarkBit();
else if (allocateBitSet(p))
free(p);
p += length(p);
}
So how do we find the beginning of the block?
– Can use balanced tree to keep track of all allocated blocks
where the key is the location
– Balanced tree pointers can be stored in header (use two
additional words)
head
left
Memory-Related Bugs
•
•
•
•
•
•
•
data
size
right
Dereferencing Bad Pointers
• The classic scanf bug
Dereferencing bad pointers
Reading uninitialized memory
Overwriting memory
Referencing nonexistent variables
Freeing blocks multiple times
Referencing freed blocks
Failing to free blocks
scanf(“%d”, val);
Page 13
Reading Uninitialized Memory
Overwriting Memory
• Assuming that heap data is initialized to zero
• Allocating the (possibly) wrong sized object
int **p;
/* return y = Ax */
int *matvec(int **A, int *x) {
int *y = malloc(N*sizeof(int));
int i, j;
p = malloc(N*sizeof(int));
for (i=0; i<N; i++) {
p[i] = malloc(M*sizeof(int));
}
for (i=0; i<N; i++)
for (j=0; j<N; j++)
y[i] += A[i][j]*x[j];
return y;
}
Overwriting Memory
Overwriting Memory
• Not checking the max string size
• Off-by-one error
int **p;
char s[8];
int i;
p = malloc(N*sizeof(int *));
gets(s);
/* reads “123456789” from stdin */
for (i=0; i<=N; i++) {
p[i] = malloc(M*sizeof(int));
}
• Basis for classic buffer overflow attacks
– 1988 Internet worm
– Modern attacks on Web servers
– AOL/Microsoft IM war
Page 14
Overwriting Memory
Overwriting Memory
• Referencing a pointer instead of the object it
points to
• Misunderstanding pointer arithmetic
int *search(int *p, int val) {
int *BinheapDelete(int **binheap, int *size) {
int *packet;
packet = binheap[0];
binheap[0] = binheap[*size - 1];
*size--;
Heapify(binheap, *size, 0);
return(packet);
}
while (*p && *p != val)
p += sizeof(int);
return p;
}
Referencing Nonexistent Variables
Freeing Blocks Multiple Times
• Forgetting that local variables disappear when a
function returns
• Nasty!
x = malloc(N*sizeof(int));
<manipulate x>
free(x);
int *foo () {
int val;
return &val;
}
y = malloc(M*sizeof(int));
<manipulate y>
free(x);
Page 15
Failing to Free Blocks
(Memory Leaks)
Referencing Freed Blocks
• Evil!
• Slow, long-term killer!
x = malloc(N*sizeof(int));
<manipulate x>
free(x);
...
y = malloc(M*sizeof(int));
for (i=0; i<M; i++)
y[i] = x[i]++;
foo() {
int *x = malloc(N*sizeof(int));
...
return;
}
Failing to Free Blocks
(Memory Leaks)
Dealing With Memory Bugs
• Freeing only part of a data structure
struct list {
int val;
struct list *next;
};
foo() {
struct list *head =
malloc(sizeof(struct list));
head->val = 0;
head->next = NULL;
<create and manipulate the rest of the list>
...
free(head);
return;
}
•
Conventional debugger (gdb)
– Good for finding bad pointer dereferences
– Hard to detect the other memory bugs
•
Debugging malloc (CSRI UToronto malloc)
– Wrapper around conventional malloc
– Detects memory bugs at malloc and free boundaries
• Memory overwrites that corrupt heap structures
• Some instances of freeing blocks multiple times
• Memory leaks
– Cannot detect all memory bugs
• Overwrites into the middle of allocated blocks
• Freeing block twice that has been reallocated in the interim
• Referencing freed blocks
Page 16
Dealing With Memory Bugs (cont.)
Disclaimer
• Parts of the slides were developed by the
course text authors: Dave O’Hallaron and
Randy Bryant. The slides are intended for
the sole purpose of instruction of computer
organization at the University of Rochester.
All copyrighted materials belong to their
original owner(s).
• Binary translator (Atom, Purify)
– Powerful debugging and analysis technique
– Rewrites text section of executable object file
– Can detect all errors as debugging malloc
– Can also check each individual reference at runtime
• Bad pointers
• Overwriting
• Referencing outside of allocated block
• Garbage collection (Boehm-Weiser Conservative GC)
– Let the system free blocks instead of the programmer.
Page 17